Rainfall is a critical climatic variable that significantly impacts various sectors, including agriculture and disaster management. Due to its high variability, accurate forecasting is essential to reduce uncertainty and support regional planning. While the Autoregressive Integrated Moving Average (ARIMA) is a standard forecasting method, the inclusion of meteorological factors such as temperature and humidity through the Seasonal Autoregressive Integrated Moving Average with Exogenous Variables (SARIMAX) approach can potentially enhance model precision. This study aims to produce a robust model to predict monthly rainfall in Makassar City for the next 12 months. The research employed a quantitative approach using secondary data from the Paotere Makassar Maritime Meteorological Station, covering the period from June 2015 to May 2025. The methodology involved splitting data into training and test sets, conducting stationarity tests via Yeo-Johnson transformation and Augmented Dickey-Fuller (ADF) tests, and determining seasonal and non-seasonal orders. Preliminary results indicated that the rainfall data exhibited significant fluctuations and required transformation to achieve stationarity in variance. Several candidate models were evaluated for parameter significance and residual white noise. The analysis identified that the SARIMAX (0,1,1)(0,1,1)[12] model is the best for modelling rainfall in Makassar City as it meets the assumptions of white noise and normality of residuals. It is concluded that this model, incorporating temperature and humidity as exogenous variables, provides a systematic framework for rainfall prediction. For future research, it is recommended to use this model for prediction provided that exogenous data for the desired period is available.
Copyrights © 2026